{"id":"https://openalex.org/W2886076728","doi":"https://doi.org/10.18653/v1/p19-1142","title":"Improved Language Modeling by Decoding the Past","display_name":"Improved Language Modeling by Decoding the Past","publication_year":2019,"publication_date":"2019-01-01","ids":{"openalex":"https://openalex.org/W2886076728","doi":"https://doi.org/10.18653/v1/p19-1142","mag":"2886076728"},"language":"en","primary_location":{"id":"doi:10.18653/v1/p19-1142","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p19-1142","pdf_url":"https://www.aclweb.org/anthology/P19-1142.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/P19-1142.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047714914","display_name":"Siddhartha Brahma","orcid":null},"institutions":[{"id":"https://openalex.org/I4210085935","display_name":"IBM Research - Almaden","ror":"https://ror.org/005w8dd04","country_code":"US","type":"facility","lineage":["https://openalex.org/I1341412227","https://openalex.org/I4210085935","https://openalex.org/I4210114115"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Siddhartha Brahma","raw_affiliation_strings":["IBM Research, Almaden, USA","IBM"],"affiliations":[{"raw_affiliation_string":"IBM Research, Almaden, USA","institution_ids":["https://openalex.org/I4210085935"]},{"raw_affiliation_string":"IBM","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5047714914"],"corresponding_institution_ids":["https://openalex.org/I4210085935"],"apc_list":null,"apc_paid":null,"fwci":0.289,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.64234965,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1468","last_page":"1476"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9962000250816345,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/treebank","display_name":"Treebank","score":0.9617135524749756},{"id":"https://openalex.org/keywords/perplexity","display_name":"Perplexity","score":0.9537424445152283},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8372125625610352},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.7955447435379028},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6988356709480286},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.6617580652236938},{"id":"https://openalex.org/keywords/character","display_name":"Character (mathematics)","score":0.5443756580352783},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5285742878913879},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5101509690284729},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.4588308036327362},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4464591443538666},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.4462336003780365},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.4361829459667206},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.13242816925048828},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.079800546169281},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.07198527455329895},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.06821203231811523}],"concepts":[{"id":"https://openalex.org/C206134035","wikidata":"https://www.wikidata.org/wiki/Q811525","display_name":"Treebank","level":3,"score":0.9617135524749756},{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.9537424445152283},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8372125625610352},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.7955447435379028},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6988356709480286},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6617580652236938},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.5443756580352783},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5285742878913879},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5101509690284729},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.4588308036327362},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4464591443538666},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4462336003780365},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.4361829459667206},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.13242816925048828},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.079800546169281},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.07198527455329895},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.06821203231811523},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.18653/v1/p19-1142","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p19-1142","pdf_url":"https://www.aclweb.org/anthology/P19-1142.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1808.05908","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1808.05908","pdf_url":"https://arxiv.org/pdf/1808.05908","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:2886076728","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1808.05908","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.1808.05908","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1808.05908","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"},{"id":"doi:10.57702/fskxprc7","is_oa":true,"landing_page_url":"https://doi.org/10.57702/fskxprc7","pdf_url":null,"source":{"id":"https://openalex.org/S7407053387","display_name":"TIB Data Manager","issn_l":null,"issn":[],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"dataset"}],"best_oa_location":{"id":"doi:10.18653/v1/p19-1142","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p19-1142","pdf_url":"https://www.aclweb.org/anthology/P19-1142.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8100000023841858}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2886076728.pdf","grobid_xml":"https://content.openalex.org/works/W2886076728.grobid-xml"},"referenced_works_count":18,"referenced_works":["https://openalex.org/W2064675550","https://openalex.org/W2117130368","https://openalex.org/W2409027918","https://openalex.org/W2510842514","https://openalex.org/W2525332836","https://openalex.org/W2553303224","https://openalex.org/W2609130030","https://openalex.org/W2611669587","https://openalex.org/W2618854269","https://openalex.org/W2795285343","https://openalex.org/W2951672049","https://openalex.org/W2962832505","https://openalex.org/W2962964385","https://openalex.org/W2963266340","https://openalex.org/W2963347649","https://openalex.org/W2963748792","https://openalex.org/W2963938518","https://openalex.org/W2963983719"],"related_works":["https://openalex.org/W2964047924","https://openalex.org/W2963537482","https://openalex.org/W2949430256","https://openalex.org/W2799253188","https://openalex.org/W3127324795","https://openalex.org/W2128634885","https://openalex.org/W3037026620","https://openalex.org/W2118813542","https://openalex.org/W2161575515","https://openalex.org/W2963971322","https://openalex.org/W3211666610","https://openalex.org/W2301095666","https://openalex.org/W2574626143","https://openalex.org/W2403097738","https://openalex.org/W2250451826","https://openalex.org/W2946704986","https://openalex.org/W2558992190","https://openalex.org/W2970798287","https://openalex.org/W2760379373","https://openalex.org/W2970949114"],"abstract_inverted_index":{"Highly":[0],"regularized":[1,96],"LSTMs":[2],"achieve":[3],"impressive":[4],"results":[5,139],"on":[6,20,74,90,115,125],"several":[7],"benchmark":[8],"datasets":[9],"in":[10,25,45,58,146],"language":[11,134],"modeling.":[12,135],"We":[13],"propose":[14],"a":[15,99,103],"new":[16],"regularization":[17],"method":[18,71,121],"based":[19],"decoding":[21],"the":[22,26,29,33,38,52,59,75,91,126],"last":[23],"token":[24],"context":[27],"using":[28,98],"predicted":[30],"distribution":[31],"of":[32,61,109,137],"next":[34,53],"token.":[35,54],"This":[36],"biases":[37],"model":[39],"towards":[40],"retaining":[41],"more":[42],"contextual":[43],"information,":[44],"turn":[46],"improving":[47],"its":[48],"ability":[49],"to":[50,81,87,111],"predict":[51],"With":[55,102],"negligible":[56],"overhead":[57],"number":[60],"parameters":[62],"and":[63,84],"training":[64],"time,":[65],"our":[66,120],"Past":[67],"Decode":[68],"Regularization":[69],"(PDR)":[70],"improves":[72],"perplexity":[73,113],"Penn":[76,127],"Treebank":[77,128],"dataset":[78,130],"by":[79,85],"up":[80,86,110],"1.8":[82],"points":[83,89,114],"2.3":[88],"WikiText-2":[92],"dataset,":[93],"over":[94,142],"strong":[95],"baselines":[97],"single":[100],"softmax.":[101],"mixture-of-softmax":[104],"model,":[105],"we":[106],"show":[107],"gains":[108],"1.0":[112],"these":[116,138],"datasets.":[117],"In":[118],"addition,":[119],"achieves":[122],"1.169":[123],"bits-per-character":[124],"Character":[129],"for":[131],"character":[132],"level":[133],"Each":[136],"constitute":[140],"improvements":[141],"models":[143],"without":[144],"PDR":[145],"their":[147],"respective":[148],"settings.":[149]},"counts_by_year":[{"year":2019,"cited_by_count":2}],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2018-08-22T00:00:00"}